Queries a knowledge graph for grounded answers, uploads and indexes new documents into it, and unlinks the files that have gone out of date.
23actions available
Every action runs live through MCP. Nothing to build, nothing to maintain.
Ayro generates text completions from Writer's AI model for fluent, contextual responses. Your team speeds up reply drafting while maintaining professional quality.
Ayro queries your knowledge graph with customer questions to find the exact answer. Your team gives accurate, sourced responses pulled directly from your documentation.
Ayro adds PDFs and documents to Writer so it can cite your actual content. Your team ensures Ayro answers from your policies and training materials, not hallucinations.
Each card is one request a support team gets, and the Writer actions Ayro runs to close it.
Ayro sends the question to a named Writer knowledge graph and returns the answer built from the documents indexed in it, so the response traces back to material your team put there.
Ayro uploads the file, parses a PDF into text or markdown when the source is a report or a scan, and adds it to the graph so it is retrievable on the next question.
Ayro unlinks a superseded file from the graph, deletes it from Writer entirely when it should not exist anywhere, and renames or re-describes the graph as its scope changes.
Ayro reads a screenshot the customer attached to pull the error text out of it, translates a reply with control over formality and length, and searches the web when the graph has nothing.
How it works
Connect once, then just ask. There is no workflow builder to learn and nothing to maintain — Ayro reads the Writer actions it has and picks the ones a request needs.
Authorize the Writer account your team already uses — one consent screen, no API keys, no mapping tables. Ayro can only do what you granted that account, and you can disconnect it at any time.
Describe the job the way you would hand it to a teammate. Ayro maps it to the Writer actions that close it and chains as many as the request needs.
The agent runs the actions inside the conversation the customer is already in, so nobody copies data between tabs and your team can take over at any point.
Get started freeWriter · Ayro
What can you do in Writer without me setting anything up?
These, right away:
Great, let's start with the first one.
FAQ
Setup, permissions, and the limits of what Ayro can do inside Writer.
That depends which action Ayro calls. Ask Question to Knowledge Graph builds its answer from the files indexed in that graph, while Text Generation and Chat Completion are open-ended model calls with no graph behind them. Point Ayro at the graph action when provenance matters.
All 23 actions your agent can call on Writer, straight from the live MCP connection.
Add a file to a knowledge graph for indexing and retrieval.
Analyze images using Writer's vision capabilities.
Send a question to the knowledge graph and retrieve the answer.
Generate chat-based completions.
Create a new knowledge graph.
Delete a file by its ID.
Delete a knowledge graph by its ID.
Detect whether content was generated by AI.
Download the binary content of a file.
Retrieve detailed information about a specific file by its ID.
List all no-code agent applications.
List all uploaded files.
Retrieve a list of knowledge graphs.
List all available language models.
Extract medical entities and concepts from unstructured clinical text.
Parse and extract text content from a previously uploaded PDF file.
Remove a file from a Knowledge Graph.
Retrieve a knowledge graph by its ID.
Generate text based on a given prompt.
Translate text from one language to another with support for formality, length control, and profanity masking.
Update an existing knowledge graph.
Upload a file to Writer.
Perform a web search.
Same connection, same setup. Pick the next one your team already uses.
Reads the available voice catalogue, returning each voice_id, name, gender, and language, and narrows it to one language code on request.
Searches the toolkit catalogue for the right action, walks users through authorising accounts, executes tools singly or in batches, and enables triggers.
Speaks written answers aloud with a chosen voice and reports whether an avatar video render has finished and where the file sits.
Searches the live web with date and domain filters, returns sourced answers or structured JSON, and fetches a page as markdown.
Runs a template against supplied data, reports app job status and results, and manages the webhook subscriptions behind those jobs.
Registers and announces agents, discovers peers by capability, opens and answers collaboration requests, and checks SwarmScore and trust records.

Authorize the account and your agent has all 23 actions from the first conversation.
